<<

Anticipated acquisition by Railway Limited of Thameslink, Southern and Great Franchise

ME/6470-14

The CMA’s decision on reference under section 33(1) given on 11 September 2014. Full text of the decision published on 7 November 2014.

Please note that the square brackets indicate figures or text which have been deleted or replaced in ranges at the request of the parties for reasons of commercial confidentiality.

Summary

1. Govia Limited (Govia) is a joint venture between Go-Ahead Holding Limited and (UK) Limited (together with their group companies referred to as ‘the Parties’), which are both operators of public transport services in the UK. It is the parent company of Limited (GTRL), an operating company incorporated to bid for and operate the Thameslink, Southern and Great Northern Rail Franchise (TSGN or the Franchise).

2. TSGN comprises two existing passenger rail franchises and parts of a third, which together will form the TSGN franchise. TSGN covers routes across , Cambridgeshire, Greater , , , , Norfolk, and East and . The (DfT) awarded the Franchise to GTRL on 23 May 2014. The existing Thameslink and Great Northern and Southern franchises carry a combined 273 million passenger journeys per year, employ around 6,500 people and generate annual passenger revenues of GBP 1.3 billion.

3. The Competition and Markets Authority (CMA) has assessed the Franchise under the merger control provisions of the Enterprise Act. This Franchise qualifies for investigation under those provisions since GTRL will acquire control of the Franchise by virtue of section 66(3) of the and the UK revenues of the Franchise exceed £70 million. The CMA therefore considers that it is or may be the case that arrangements are in progress or in contemplation which, if carried into effect, will result in the creation of a relevant merger situation pursuant to section 23(2) of the Enterprise Act (the Act).

1 4. The parties overlap in the provision of public passenger transport services, ie bus (Go-Ahead) and rail (Govia and the Franchise). The CMA has considered competition on a flow-by-flow basis on those flows where the parties overlap.

5. The TSGN franchise agreement is a management contract in which the operator manages the delivery of rail services on the Franchise on behalf of the DfT. Under this contract, the DfT takes almost all the rail fare revenues and the operator earns a management fee. The contract imposes significant constraints on the operator’s freedom to determine the commercial parameters of the Franchise, such as ticket prices. This is different from standard rail franchise awards where greater commercial freedom is afforded to the operator.

6. The CMA has considered whether the Merger could lead to an increase in prices or worsening of services on either the overlapping non-Franchise services already operated by the Parties (since customers lost as a result would switch to the Franchise) or on the Franchise (since lost customers would switch to Govia’s other services). In each case the CMA has consid- ered whether the Parties have the incentive and ability to increase prices or degrade services. The CMA considers that while the Parties may in principle have the ability to increase prices or degrade services on some of their overlapping non-Franchise services, it will have very limited incentive to do so since it will make little, if any, additional revenue from customers switching to the Franchise as a result. The CMA therefore believes that there is little or no incentive for the Parties’ to worsen the services or prices of their overlapping non-Franchise services.

7. Since the Parties would only lose minimal revenue in losing passengers on the TSGN Franchise, they may have an incentive to attract customers away from the Franchise in the hope of capturing them on their competing services. However, the CMA considers that Govia is prevented from increasing prices on the Franchise by the contractual obligations under the TSGN franchise agreement and DfT involvement and monitoring mechanisms. In addition, Govia is subject to particularly strict quality controls which substantially limit Govia’s ability to degrade the service, as well as a financial penalty and incentive structure linked to quality targets.

8. The CMA considers that these constraints, taken together, are sufficient to ensure that no realistic prospect of a substantial lessening of competition will arise as a result of the Merger.

9. This merger will therefore not be referred under section 33(1) of the Act.

2 Assessment

Parties

10. GTRL is a operating company incorporated for the purpose of bidding for and operating the TSGN franchise. GTRL is a wholly owned subsidiary of Govia.

11. Govia holds, through its subsidiary train operating companies (TOCs), the current contracts to run the Southern,1 Southeastern,2 and London Midland3 rail franchises. Govia is a joint venture (JV) between 65% shareholder Go- Ahead and 35% shareholder Keolis. Under the relevant shareholders agreement Go-Ahead may [] and Keolis may []. []

12. Go-Ahead is a UK public transport operator. It had UK turnover in the year ending June 2013 of approximately £2.6 billion.

13. Keolis is an international operator of passenger transport services, ultimately owned by SNCF, the French state . Its UK turnover in the year ending December 2013 was £[], and its worldwide turnover was £4 billion.4 In the UK, as well as its shareholding in Govia, Keolis has a 45% shareholding in FirstKeolis which currently runs the TransPennine Express rail franchise in northern . It also has a 70% shareholding in a JV with Amey that was recently awarded the contract to operate the .

14. The TSGN Franchise covers routes across Bedfordshire, Cambridgeshire, Greater London, Hampshire, Hertfordshire, Kent, Norfolk, Surrey and East and West Sussex. The DfT is the franchising authority for the Franchise. Under the TSGN Franchise Agreement, GTRL acquires the right to operate certain passenger rail services, specifically:

(a) Thameslink and Great Northern – currently owned and operated by Limited (operated as FCC). These services will transfer to the TSGN Franchise on 14 September 2014.

1 Govia has operated the Southern franchise since 20 September 2009 through its TOC Southern Railway Limited (and from 2001 through its TOC New Southern Railway Limited). The Southern franchise will end and transfer to the TSGN franchise on 26 July 2015. 2 Govia has operated the Southeastern (or Integrated Kent) franchise since 1 April 2006 through its TOC London and South Eastern Railway Limited. The Southeastern franchise is due to be re-franchised in 2018, and some of its routes will transfer to the TSGN franchise in December 2014 and January 2018. 3 Govia has operated the franchise since 11 November 2007 through its TOC London and Birmingham Railway Limited. The London Midland franchise is due to be re-franchised in 2017. 4 €[] UK turnover, and €4,920.7 million worldwide turnover.

3 (b) Southern (including ) – currently owned and operated by Southern Railway Limited, a wholly owned subsidiary of Govia. These services will transfer to the TSGN Franchise on 26 July 2015.

(c) Parts of Southeastern – currently owned and operated by London and South Eastern Railway Limited, a wholly owned subsidiary of Govia. Parts of the Southeastern services will transfer to the TSGN Franchise on 21 December 2014 and 2 January 2018.

Transaction

15. On 23 May 2014 the DfT announced that GTRL was the successful bidder for the TSGN rail franchise. On 11 June 2014 (after a statutory standstill period), the Franchise Agreement between the Secretary of State for Transport and GTRL was executed, confirming the award of the TSGN franchise to GTRL. The Franchise will commence on 14 September 2014 and has been awarded for an initial period of seven years, ending on 19 September 2021.

16. The rail services that are being transferred to the Franchise have existing turnover and goodwill and GTRL will acquire the assets, staff, rights and liabilities necessary to operate the TSGN Franchise. The existing Thameslink and Great Northern and Southern franchises carry a combined 273 million passenger journeys per year, employ around 6,500 people and generate annual passenger revenues of GBP 1.3 billion.

Jurisdiction

17. The award of a rail franchise constitutes an acquisition of control of an enterprise by virtue of section 66(3) of the Railways Act 1993. Govia and the Franchise will therefore cease to be distinct.

18. Govia and its TOCs (including GTRL) are jointly controlled by Go-Ahead and Keolis. Under the relevant shareholders’ agreement, each of []. Go-Ahead and Keolis have therefore gained joint control, through Govia, of TSGN.

19. Therefore, as a result of the Merger, the enterprises of Govia and the TSGN Franchise will cease to be distinct. The UK turnover of the TSGN Franchise exceeds £70 million, so the turnover test in section 23(1)(b) of the Act is satisfied. The CMA therefore believes that it is or may be the case that arrangements are in progress or in contemplation which, if carried into effect, will result in the creation of a relevant merger situation.

20. The Merger meets the thresholds under Council Regulation (EC) 139/2004 (the EC Merger Regulation) for review by the EU Commission (the

4 Commission). The parties submitted a reasoned submission to the Commission on 10 July 2014 requesting pre-notification referral to the CMA under Article 4(4) of the EC Merger Regulation. The CMA informed the Commission that it agreed with the referral request and considered the Merger capable of being reviewed in the United Kingdom under the Act. On 4 August the Commission announced its decision to refer the Merger to the CMA for review.5 The statutory deadline for a decision by the CMA under section 34A of the Act is therefore 7 October 2014.6

Frame of reference

21. The CMA considers that market definition provides a framework for assessing the competitive effects of a merger and involves an element of judgement. The boundaries of the market do not determine the outcome of the analysis of the competitive effects of the merger, as it is recognised that there can be constraints on merging parties from outside the relevant market, segmentation within the relevant market, or other ways in which some constraints are more important than others.7

22. The parties overlap in the provision of public passenger transport services, including bus (Go-Ahead) and rail (Govia and the Franchise).

Product frame of reference

23. The Parties submit that the relevant product market for assessing the Merger is the supply of public passenger transport services, in particular rail and bus services.

24. In line with previous decisions the CMA has in the present case considered, on a cautious basis, public transport as a separate product market to private transport given the limited substitutability.8 There are various modes of public transport (rail, bus, coach etc) that are available to passengers wishing to make a particular journey.9 The Competition Commission (CC) has previously

5 Case M.7146 – GOVIA/THAMESLINK, SOUTHERN AND GREAT NORTHERN PASSENGER RAIL FRANCHISE. 6 Upon a referral back from the Commission, the CMA has, under section 34A of the Act, a maximum of 45 working days to inform the merger parties of the CMA’s SLC decision, beginning on the working day after receipt of the Commission’s referral decision. However, the CMA will generally try to decide the case within the 40- working-day timetable used for ‘standard’ mergers. See Mergers: Guidance on the CMA’s Jurisdiction and Procedure (CMA2, January 2014), paragraph 18.22. 7 Merger Assessment Guidelines (CC2/OFT1254, September 2010), paragraph 5.2.2. The Merger Assessment Guidelines have been adopted by the CMA (see Annex D to Mergers: Guidance on the CMA’s Jurisdiction and Procedure). 8 See, for example, Competition Commission, Review of methodologies in transport inquiries (2007); ME/6226-13 Completed acquisition by Passenger Services Limited of the remainder of the entire share capital of Centrebus Holdings Ltd. 9 ME/2238/06 – London and South Eastern Railway / Integrated Kent Rail Franchise (28 March 2006).

5 noted that there are a range of factors influencing their choice including journey time and cost and these factors may differ depending on the exact journey to be made.10 The CMA has in this case considered the supply of all public transport services as a starting point for the purpose of assessing the competitive effect of the Merger but notes that the precise product scope varies on a flow-by-flow basis.

25. The CC has previously distinguished leisure travel from business travel and commuting due to different sensitivities in price, journey time and duration.11 This is taken into account where relevant in the CMA’s competitive assess- ment of the present case to the extent that different ticket types (ie regulated or non-regulated) have been considered,12 but the CMA does not consider it necessary to conclude on whether leisure and business passengers should be considered separately given that no competition concerns arise on any basis.

Geographic frame of reference

Flow level competition

26. As passengers travel between a specific point of origin to a specific destin- ation, the demand is therefore for travel between two points. Consistent with previous cases, the CMA assesses the impact of the transaction on a flow-by- flow basis, where a ‘flow’ is defined as a connection between two specific points. The CMA also considers whether nearby flows on other routes may be in the same market to the extent that they are near enough to be substitutable from a passenger perspective.

27. While the CMA does not find it necessary to conclude on market definition given that no competition concerns arise on any basis, the CMA uses the following definitions of overlaps in this case:

Bus-on-rail overlaps

28. The National Travel Survey (NTS) by the DfT13 suggests that:

(a) Among bus passengers making a journey from their home, 80% live within 417 metres of their nearest bus stop.14

10 Competition Commission, Review of methodologies in transport inquiries (2007). 11 ibid. 12 Although the CMA notes that some passengers may switch ticket types without their journey type changing. 13 See National Travel Survey statistics. 14 DfT estimate.

6 (b) Among rail passengers travelling less than 12 miles,15 an estimated 80% of train journeys involved walking less than 805 metres to catch the train.16

29. Together this suggests that for bus-rail overlaps for short journeys in urban areas, an appropriate measure of overlap is 1,200 metres (where the approximate 400-metre catchment area around a bus stop abuts with the approximate 800-metre catchment area around a ).17 That is, flows for which the beginning and end points are within 1,200 metres of one another are considered by the CMA to be overlapping. The CMA has not needed to consider whether there are specific reasons for catchment areas to be different in the current case (for example, a smaller overlap measure may be appropriate in areas of London due to public transport density) given that no competition concerns arise on any basis.

30. For other (bus-rail overlap) journeys, the CMA in this case has considered as overlapping any two bus and rail flows that join the same two ‘settlements’. Consistent with previous cases, it has used this approach for flows that join: two rural settlements, a commuter town or village and a large urban centre (such as London), or two settlements linked by a long-distance service (approximately 50 miles or more).18

Rail-on-rail overlaps

31. For rail-on-rail overlaps, overlaps could occur in the following ways:

(a) journeys between the same two stations;

15 This restriction is relevant when considering the distance bus passengers will walk to train stations. Around 95% of local bus journeys are less than 12 miles (analysis of NTS data submitted by [], a competitor of Govia). Considering shorter train journeys is therefore consistent with assessing overlaps for short distance bus services. 16 Analysis of NTS data submitted by [], a competitor of Govia. The CMA notes that the actual figure could be larger than 805 metres due to possible underreporting of walks to train stations in the NTS data. However, 805 metres is consistent with the distance used in previous rail franchise cases (eg A report on the proposed acquisition by FirstGroup plc of the Scottish Passenger Rail Franchise currently operated by ScotRail Railways Limited, CC report 2004). It is also similar to measures used for other purposes such as ’s Public Transport Accessibility Levels, which uses a maximum walking catchment area around rail, underground and light rail service stations of 960 metres. 17 DfT analysis of NTS data suggests that for passengers that used both rail and bus during a recorded travel week (for trips of 12 miles or under, starting from the respondent’s home), 80% of individuals estimated the walk time to their nearest station to be 25 minutes or less. This is equivalent to an estimated distance of 2,083 metres (assuming a walking speed for an average person of 5 km per hour). For the purposes of identifying bus-rail overlaps, the CMA has not used this figure, because over this distance some of these individuals may not have walked to the station but used another mode of transport such as a bus. This is not a material issue in the case of the 417-metre distance for bus stops (noted in paragraph 28(a)), as this distance is sufficiently short to expect that passengers would have walked this distance. The CMA has therefore used this 417-metre figure for bus stops. 18 The DfT has defined long-distance journeys using this measure. For example, see National Travel Survey 2013 (p9).

7 (b) journeys between the same two towns, even where different stations in the same town are used;19 and

(c) journeys to a main commuter city such as London from two alternative stations in nearby towns or villages.

32. The CMA has considered all such rail-on-rail overlaps identified by the Parties in its competitive assessment.

Counterfactual

33. The CMA assesses a merger’s impact relative to the situation that would prevail absent the merger (that is, the counterfactual). In practice, the CMA generally adopts the pre-merger conditions of competition as the counter- factual against which to assess the impact of the merger. However, the CMA will assess the merger against an alternative counterfactual where, based on the evidence available to it, there is a realistic prospect of a counterfactual that is more competitive than the pre-merger conditions.20

34. In rail franchise cases, the pre-merger situation cannot be the appropriate counterfactual, as the existing rail franchise is coming to an end and a new franchise must be awarded to one of the short-listed bidders. The CMA therefore treats the appropriate counterfactual to the Merger as the award of the franchise to a firm that raises no competition concerns or where any competition concerns could be remedied through undertakings in lieu of a reference to phase 2.21

Competitive assessment

Horizontal unilateral effects

Competition for the market

35. The CMA considered the effect on competition for the market (ie competition for rail franchises) for which the parties’ national market shares are relevant. The parties estimate that the award of the Franchise increases Govia’s market share to [20–30]% with an increment of [5–10]% which is at a level that will not often give the CMA cause for concern.22 In addition, the Merger does not reduce the number or identity of likely bidders for future franchises.

19 In the present case for example, Maidstone West, Maidstone East and Maidstone Barracks. See Competition Commission report on Group plc and the Greater Anglia franchise, paragraph 4.3. (November 2004) 20 See Merger Assessment Guidelines, paragraph 4.3.5 et seq. 21 ibid, paragraphs 4.3.28 and 4.3.29. 22 ibid, paragraph 5.3.5.

8 The CMA does not therefore consider that there is a realistic prospect of a substantial lessening of competition for rail franchises as a result of the Merger and has therefore not considered competition for the market further in its decision.

Flow-by-flow overlaps

36. The extent of the overlaps between the Franchise and the parties will vary during the lifetime of the Franchise given the different stages at which parts of the Franchise are being transferred to GTRL (set out in paragraph 14 above). In summary, the overlap services are set out in Table 1 below.

TABLE 1 Summary of Govia’s overlaps with the Franchise

Govia’s services From September 2014 From July 2015 From December 2014 From January 2018

Southern rail Two routes (parts of franchise which are in London).*

Southeastern rail One route going outside Three route overlaps on a franchise London. station-to-station basis going outside London. Multiple routes inside London. One route (on a wider overlap basis) going outside of London.

Multiple routes inside London.

London Midland One route (from July One route going outside rail franchise 2015) going outside London. London.

Transport for Overlaps on multiple London routes contracted bus services operated by Go-Ahead

Go-Ahead bus 6 rail routes (all direct) 22 rail routes with services operated direct overlaps outside Greater London 73 indirect routes which have overlaps on a few flows (typically one or two flows)

*These services will become part of the TSGN franchise as of July 2015 and would therefore have been operated by the TSGN Franchisee in any event.

37. It is also relevant to note that the current expiry dates for the Southeastern and London Midland franchises operated by Govia (and which overlap geographically with TSGN in some places) are October 2014 and April 2016 respectively. For the purposes of their overlap analysis, the Parties assumed that these are extended through direct awards, as detailed in the DfT’s recently published franchise programme, to June 2018 and June 2017.

9 38. The CMA has considered whether the Merger could lead to an increase in prices or worsening of services on either the overlapping non-Franchise services already operated by the Parties (since customers lost as a result would switch to the Franchise) or on the Franchise (since lost customers would switch to Govia’s other services). In each case the CMA has consid- ered whether the Parties have the incentive and ability to increase prices or degrade services.

39. The TSGN franchise agreement differs from most rail franchise awards in that it has been structured as a management contract whereby the DfT takes almost all the rail fare revenue.23 The incentive of the TSGN operator to degrade or increase price of any of its other overlapping public transport services (‘non-Franchise’ services) may therefore be limited: almost all additional revenue from rail fares due to passengers switching to the Franchise would go to the DfT. Since the TSGN operator does not benefit in the same way as other franchises from increased rail fare revenues,24 the DfT has imposed greater restrictions on the TSGN franchise which may also reduce the ability of the operator to raise prices or degrade services. This is considered in more detail below.

Price increase or degradation of services on non-Franchise services

40. The Parties operate three types of overlapping non-Franchise services: Transport for London (TfL) contracted bus services; the Southern, Southeastern and London Midland rail franchises; and non-TfL bus services (ie, outside London).

41. In regards to the TfL contracted bus services, TfL told the CMA that it sets the prices, frequencies and other aspects of service quality, with Go-Ahead’s activities limited to operating the buses under a management agreement. The CMA therefore considers that, in the specific circumstances of this case and taking account of the specific contractual arrangements, Go-Ahead has no ability to degrade its TfL contracted bus services to any material extent and will not consider these services further.

42. Information obtained from the parties and the DfT indicates that there are also some constraints on Govia’s ability to alter service quality or prices on the other rail franchises it operates, given the regulation and monitoring provided

23 The operator will retain revenue from ‘ancillary services’, for example commission from the physical sale of tickets, car parking and the sale of retail and advertising space. The CMA understands that together these services account for approximately []% of the Franchise’s overall revenue. 24 The TSGN Franchise, unlike other franchises, is structured as a management contract in order for the operator to focus on the planned changes to services and infrastructure such as the redevelopment of .

10 by the DfT for rail franchises. However, the CMA considers that the Southern, Southeastern and London Midland rail franchises do contain flexibility, in particular to adjust unregulated fares. In principle, therefore, Govia may be able to degrade its overlapping rail services if it wanted passengers to switch to the Franchise.

43. The CMA also understands that in regards to its bus services outside London, Go-Ahead is likely to have significant flexibility to adjust fares and services including with regard to frequencies and other parameters of quality.

44. The Parties may therefore have the ability to increase prices or degrade services on some of their overlapping non-Franchise services. However, the CMA considers that it will have very limited incentive to do so since they will make little, if any, additional revenue from passengers switching to the Franchise as a result.

45. This is the case in particular because the Franchise is structured as a man- agement agreement and Govia is remunerated through a management fee which is not linked to passenger numbers or fare revenues. The margin achieved by the operator is the difference between the management fee and the costs incurred in operating the Franchise;25 the operator will therefore not earn any margin on either revenue or profit). Govia passes the large majority of fare revenues back to the DfT having deducted its management fee.26

46. The CMA therefore considers that the Parties have no material incentive to attract passengers from overlapping non-Franchise services to the Franchise by worsening services or prices on these services.

Degradation of services or price increase on the Franchise

47. The CMA considered whether Govia would have the ability to increase prices or degrade services on the TSGN Franchise. The CMA notes that, given that the management contract structure means that Govia would only lose minimal revenue in losing passengers on the TSGN Franchise, the incentives to divert passengers away from the Franchise in the hope of capturing them on their overlapping services is likely to be greater than in other rail franchise operations. In addition, Govia has a prima facie incentive to reduce quality irrespective of the existence of overlapping flows since Govia would be able to reduce costs without losing revenue. Below the CMA considers Govia’s ability to alter services and price separately.

25 See definition of ‘Estimated Profit Stream’ in the Franchise Agreement dated 11 June 2014. 26 Schedule 8.1, Franchise Payments of the Franchise Agreement dated 11 June 2014.

11 Degradation of services on the Franchise

48. The CC has in the past concluded that there may be little opportunity for operators to degrade quality on rail services (such as frequency or punc- tuality),27 since service quality on particular flows is constrained by regulation, the franchise agreement, and the difficulty in degrading a particular flow without also affecting other flows on the same route where there is no overlap.

49. The CMA considers that it is possible to regulate the parameters of quality relatively effectively in rail franchises in so far as at least certain parameters of quality (such as frequency and delays) are generally quantifiable and measur- able. In the present case, the parties submitted, and the DfT confirmed, that the TSGN franchise agreement contains particularly strict quality controls which substantially limit Govia’s ability to degrade the service. This is reflected, for instance, by the fact that the margin earned by the Parties in running the Franchise is the difference between its management fee received and the costs incurred in operating the Franchise. The DfT has taken significant steps to prevent this contractual structure from providing the operator with a prima facie incentive to reduce costs by degrading the quality of service. In particular, the DfT has specified the frequency and service times of the Franchise as well as certain binding capacity requirements (number, size and type of carriages).

50. In addition, the DfT has put in place a system of penalties and bonuses, and supporting monitoring systems, to ensure that quality is maintained. The operator’s performance is measured in regards to factors including delay, cancellations and a passenger experience metric looking at elements such as cleanliness, rubbish collection and security.

51. The TSGN Franchise operator will be able to earn additional financial incentives for exceeding its target benchmarks with regard to the factors listed above (and others). The CMA considers that this will provide a significant incentive for the operator to hit or exceed its targets. Moreover, the penalty for missing the targets will [] acting as a significant disincentive. In addition, if the franchise falls short of its targets, it could be in breach of the franchise agreement and could therefore ultimately risk the agreement being termin- ated. The DfT, through a dedicated TSGN Franchise Management Team with access to a significant amount of information on the Franchise, will monitor the targets across the whole Franchise, including all of the Franchise flows

27 In its review of transport methodologies the CC noted that ‘The scope to reduce services on rail franchises is also limited by regulation. In past inquiries, the CC has reached the view that there is negligible scope to reduce service levels on regulated franchises; franchise agreements often specify investments and service improvements and TOCs have to comply with a range of service quality standards.’ (Paragraph 66.)

12 where the Franchise overlaps with Govia’s non-Franchise services. The CMA therefore considers that degrading services at the route level would put Govia at risk of missing its franchise-wide targets.

52. For the reasons set out above, and in particular the specific nature of the TSGN Franchise, the CMA therefore does not consider that Govia has the ability to degrade service quality on the Franchise.

Price increases on the Franchise

53. Based on information provided by the Parties and the DfT, the CMA has identified a number of factors that limit the ability of Govia to increase prices on the TSGN flows which overlap with its other services.

54. Govia submitted that DfT exerts control over fares through the Franchise Agreement.28 It noted that Govia is required to maintain a ‘bid fares policy’ that sets out how the TSGN franchisee will manage every type of fare. Govia considered that it must obtain approval from the DfT for any fares or fare policy change that is not profit maximising to the Franchise. Govia also stated that it must provide any information that the DfT may reasonably require (whether directly or indirectly relevant to fares). In summary, it was of the view that the commercial reality of the Franchise was such that it expects to work in partnership with the DfT to fulfil its agency-like responsibilities with regards to fare setting.

55. The CMA considers that Govia faces some restrictions on its ability to set prices on the Franchise that apply to all rail franchises. First, the CMA notes that all rail fares within London are zonal fares so rail fares within or across the same zone or zones are set at the same level through annual negotiations between London train operators.29 This significantly limits the ability of Govia to raise TSGN fares within London.

56. Further, the Franchise is also subject to the rail fare regulations that apply to all franchises in England and . The Government applies a regulated fares system to restrict the increases permitted on certain fares the terms of which form part of the contract with each franchise operator. In relation to TSGN, the regulation applies across two ‘baskets’ of fares, commuter and other regulated fares as specified, in relation to which Govia may only make increases:

28 As set out in Schedule 5.8 to the franchise agreement. 29 This is set out in the ATOC Zonal Fares Scheme. The outcome of these negotiations is put to a vote as part of this scheme, where at least 70% of the votes cast must be in favour of the resolution. The number of votes each participant in the scheme has is pro rata to the income it receives from zonal fares.

13 (a) to the average fare within the basket by a maximum of RPI plus 1%;30 and

(b) to any individual fare within the basket by a maximum of RPI plus 3%.31

57. Nevertheless, the CMA notes that Govia could raise regulated fares on overlapping flows to the maximum permitted level while reducing regulated fares on others to maintain the average. Such individual increases may remain a concern as there may be certain overlapping flows which are profit- able enough to offset the deduction on other flows. In addition, in the present case the potential average decreases would not be to Govia’s detriment since revenue goes to the DfT. This may mean that this fare regulation alone may not be sufficient to rule out concerns.

58. However, there are other aspects of the TSGN Franchise specifically, as noted by Govia and confirmed by the DfT, that the CMA considers place very significant restrictions on Govia’s ability to increase prices notwithstanding the limits to the general price regulation. Under the franchise agreement, Govia must set all fares in accordance with the ‘good operator standard’ which provides, in summary, that fares must be set in a way that is profit maximising for the Franchise, operating independently of Govia’s other businesses. The DfT advised the CMA that it considers that any fare increases motivated by a wish to encourage customers on to overlapping Govia services would be likely to amount to a breach of this standard. This principle applies to unregulated as well as regulated fares.

59. In addition, as noted in paragraph 54 above, as part of the franchise bidding process Govia provided the DfT with a bid fares policy setting out a clear and quantified description of the creation of regulated and unregulated fares. This is now part of the franchise agreement that Govia is bound by. The bid fares policy notes the assumption that fares within the regulated fares baskets will rise by RPI+1%, unless otherwise directed by the DfT.32 This also states that in addition to the restriction on regulated fares, average increases for unregulated fares will be RPI plus 1%.33 Any alterations to the policy must be agreed by the DfT.

60. Finally, the DfT informed the CMA that it has in place detailed monitoring plans to ensure that it can detect any action by Govia that is in breach of its contract (including the bid fares policy and good operator standard). This will include a team dedicated to the consideration of fare changes. The CMA has

30 Paragraph 2.3 of the bid fares policy. 31 ibid and DfT, Rail Fares and Ticketing: Next Steps, October 2013. The government also announced on 7 September 2014 that no regulated fares will rise by more than inflation in 2015. 32 Bid fares policy, paragraph 2.4. 33 ibid, paragraph 3.15.

14 shared with the DfT the detailed overlaps identified by Govia (summarised in Table 1 above) and the DfT has confirmed that it has put in place a dedicated franchise management team that has a very significant level of direct control over the types, and price levels, of rail ticketing. In particular, the TSGN Franchise Management Team will monitor closely proposed fares in those areas where overlaps have been identified.

61. The CMA notes that given these provisions, there is a significant risk that the DfT would find Govia to be in breach of the Franchise Agreement if Govia was to raise fares on the Franchise to attract passengers on to its overlapping services. Such a breach would not only have short-term financial conse- quences for Govia (such as the instigation of a remedial plan under Schedule 10 of the Franchise Agreement or ultimately statutory penalties under the Railways Act 1993 but Govia submitted that this may also [].

62. For the reasons set out above, the CMA therefore considers that the specific features of the TSGN mean that Govia is prevented by its contractual obli- gations under the franchise agreement and DfT involvement and monitoring mechanisms from increasing prices on fares to the extent that it could attract passengers away from the Franchise and on to its overlapping services.

Conclusion on horizontal unilateral effects

63. As set out above, the CMA considers that the Merger does not give rise to a realistic prospect of a substantial lessening of competition in respect of Govia worsening the services or prices of its overlapping non-franchise services, as there is no material incentive for Govia to do this given that DfT takes almost all TSGN revenues.

64. In addition, taking together the range of constraints and monitoring to which Govia is subject in its operation of TSGN, the CMA considers that there is also no realistic prospect of a substantial lessening of competition in respect of Govia’s degradation (or lack of improvement) of services or implementation of price increases on the TSGN franchise, as there is no material ability for Govia to do this, given in particular the restrictions faced by Govia under the franchise agreement.

65. Accordingly, the CMA has found that the Merger does not give rise to a realistic prospect of a substantial lessening of competition as a result of horizontal unilateral effects in relation to the overlapping flows.

15 Network effects in relation to multi-modal travel

66. Govia’s bus services may be complementary to the rail services it provides on the TSGN franchise, for example where passengers use bus services to get to a train station before continuing their journey by train. The CMA has considered the possibility that the Merger could result in adverse effects on competition between bus operators by enabling Govia to leverage its position as a rail operator to weaken any competition on its bus services. This includes leveraging its position in rail by offering combined bus and rail tickets restricted to its own services.34

67. When the OFT and CC have considered this theory of harm previously,35 they have noted that multi-modal ticketing can be prima facie pro-competitive but that there can be concerns under certain circumstances. Specifically, the OFT has noted that under extreme cases the adverse effects from excluding other transport operators from multi-modal ticketing could induce the exit of these other operators.

68. The main integrated train and bus ticketing scheme in , Plusbus, is managed by Journey Solutions, a partnership of bus and train operators in Great Britain. Govia, through Go-Ahead, has only one representative on Journey Solutions’ board of nine members. In addition, the TSGN Franchise Agreement includes a commitment for Govia to cooperate in the development and implementation of integrated and multi-modal fare schemes.36 Given this, the CMA considers that it would be difficult for Govia to exclude a rival bus operator from taking part in a combined rail and bus ticket scheme.

69. The CMA further notes that in general it appears difficult to marginalise a bus operator by excluding it from multi-modal tickets. This is because 95%37 of local bus journeys do not involve another mode of transport other than walking. Typically, bus operators’ revenues will therefore not be significantly dependent on multi-modal tickets. In relation to marginalising bus operators in

34 In cases where Govia is the incumbent operator on the TSGN Franchise, the relevant question is if it has leveraged its position as a rail operator to weaken competition to its bus services during this time and if so, whether this is likely to continue if that part of the TSGN Franchise is awarded to it a second time. 35 Cases where this theory of harm were considered include: Competition Commission Report on the acquisition by National Express Group plc of the Greater Anglia franchise (November 2004); OFT Case ME/1347/04 – FirstGroup plc / InterCity East Coast Franchise (21 December 2004), at paragraphs 67–69; OFT Case ME/1370/04 – Investments Limited/ plc / InterCity East Coast Franchise (21 December 2004), paragraphs 71 and 72; and OFT Case ME/2726/06 – Stagecoach / South Western (27 February 2007), paragraphs 56–61. 36 TSGN Franchise Agreement, Schedule 13, paragraph 10. 37 As noted in Journey Solutions partnership, Door-to-door by public transport, June 2009, relying on 2006 National Travel Survey data.

16 the specific areas covered by the TSGN Franchise, the CMA notes that no bus operator raised such a concern in its market testing.38

70. Given the above, the CMA considers that Govia has no ability to weaken competition in bus services by leveraging its position in rail services.

Third party views

71. The CMA contacted relevant third parties including the DfT, TfL, competitors and passenger groups. No third parties raised concerns regarding the Merger.

72. Third party comments have been taken into account where appropriate in the competitive assessment above.

Decision

73. Consequently, the CMA does not believe that it is or may be the case that the merger may be expected to result in a substantial lessening of competition within a market or markets in the United Kingdom.

74. This merger will therefore not be referred under section 33(1) of the Act.

Nelson Jung Director of Mergers Competition and Markets Authority 11 September 2014

Endnote

In relation to paragraphs 5, 39, 45, 47, 49 and footnote 24, the DfT notes that the characterisation of the TSGN Franchise Agreement as a management contract does not reflect the true nature of the Franchise Agreement, particularly as the Franchise Agreement incentivises Govia to operate the Franchise in a way that maximises revenue (in line with the ‘good operator standard’) rather than merely carry out the day-to-day operational management of the Franchise.

38 The CMA contacted a range of bus operators that competed with Go-Ahead in areas covered by the TSGN Franchise.

17